1

Longitudinal associations between formal volunteering and well-being among retired older

people: Follow-up results from a randomized controlled trial

Michelle I Jongenelisa*, Ben Jacksonb, Robert U Newtonc, and Simone Pettigrewde

a
Melbourne Centre for Behaviour Change, Melbourne School of Psychological Sciences
University of Melbourne
P: +61 03 9035 4979; Michelle.jongenelis@unimelb.edu.au

b
School of Human Sciences (Exercise and Sports Science), University of Western Australia
Ben.jackson@uwa.edu.au

c
School of Medical and Health Sciences, Edith Cowan University
r.newton@ecu.edu.au

d
The George Institute for Global Health
SPettigrew@georgeinstitute.org.au

e
School of Public Health and Community Medicine, University of New South Wales

Funding statement: This study was funded by an Australian Research Council Discovery Grant

(DP140100365).

Acknowledgements: The authors wish to thank Nicole Biagioni, Zenobia Talati, and the team of

staff and students at Curtin University and the Vario Health Clinic at Edith Cowan University for

their assistance with data collection.

 2

Longitudinal associations between formal volunteering and well-being among retired older

people: Follow-up results from a randomized controlled trial

Abstract

Volunteering has been identified as a potential mechanism for improving the psychosocial health

of older adults. Utilizing a randomized controlled trial approach, the present study assessed the

extent to which commencing volunteering can improve psychosocial health outcomes for older

people. Fully-retired Australian adults aged 60+ years (N = 445) were assessed at baseline and

allocated to either the intervention or control arms of the trial. Those in the intervention

condition were asked to participate in at least 60 minutes of formal volunteering per week for six

months. Per-protocol analyses were conducted comparing psychosocial outcomes for those who

complied with the intervention condition (n = 73) to outcomes for those who complied with the

control condition (n = 112). Those who complied with the intervention condition demonstrated

significant improvements in life satisfaction, purpose in life, and personal growth scores over a

12-month period relative to those in the control condition who did no volunteering. Findings

provide evidence of a causal relationship between commencing volunteering and improvements

in psychosocial health among older adults and indicate that encouraging participation in this

activity could constitute an effective healthy aging intervention.

Keywords: Older adults; Volunteering; Randomized controlled trial; Psychosocial health.

 3

Longitudinal associations between formal volunteering and well-being among retired older

people: Follow-up results from a randomized controlled trial

Populations around the world are aging rapidly, prompting a growing emphasis on healthy aging

to ensure increased longevity is accompanied by functional ability and psychological well-being

(World Health Organization, 2015). According to the Activity Theory of Aging (Lemon et al.,

1972), healthy aging is more likely to occur when older adults maintain their engagement in

social and productive activities because they (i) protect older adults from the greater number of

role losses that characterize older adulthood and (ii) increase life satisfaction. As a means by

which older adults can continue contributing to the community and interacting with others,

formal volunteering represents an activity that could be prescribed by health professionals as an

intervention to foster healthy aging (Pettigrew et al., 2019).

Cross-sectional and longitudinal research has found older adults’ participation in

volunteering to be associated with numerous favorable physical and psychosocial health

outcomes, including higher cognitive functioning; higher self-rated health; reduced mortality;

lower prevalence of hypertension; higher levels of life satisfaction; higher self-esteem, personal

growth, and purpose in life; lower rates of depression; greater social connectedness and social

support; and reduced loneliness (Anderson et al., 2014; Burr et al., 2015; Carr et al., 2018; Cho

et al., 2018; Greenfield & Marks, 2004; Han & Hong, 2013; Heo et al., 2017; Jenkinson et al.,

2013; Lum & Lightfoot, 2005; Parkinson et al., 2010; Pettigrew & Roberts, 2008; Pilkington et

al., 2012; Proulx et al., 2018; Tomioka et al., 2017; Wahrendorf et al., 2008). However, a major

limitation of this research is the inability to determine whether the observed health differences

between volunteers and non-volunteers are a result of their engagement in volunteering or the
 4

tendency for healthier older adults to be more likely to volunteer in the first place (Jenkinson et

al., 2013; Pettigrew et al., 2019). Results from quasi-experimental studies that have attempted to

address this limitation indicate that volunteering is associated with (i) fewer depressive

symptoms and functional limitations and (ii) greater engagement in physical activity (Hong &

Morrow-Howell, 2010; Tan et al., 2009). However, given the lack of randomization, the

potential for confounding bias remains.

Randomized controlled trials (RCTs) are best positioned to definitively establish

causality, however few appear to have been conducted to investigate the benefits of commencing

volunteering for older people, and results have been mixed (Carlson et al., 2008; Fried et al.,

2004; Jiang et al., 2020; Pettigrew et al., 2019; Rook & Sorkin, 2003; Tan et al., 2006). In a trial

that focused on the psychosocial outcomes of volunteering (Rook & Sorkin, 2003), follow-up

assessments conducted one and two years after baseline indicated that older adults randomly

assigned to volunteer for a child inpatient hospital visitation program for which they had

expressed interest were more likely than those in the comparison non-volunteer groups to form

new social ties. However, psychological health (e.g., self-esteem, depression) did not differ

between those assigned to participate in this program and those who were not.

In a similar trial that focused on the physical, cognitive, and social outcomes of

volunteering (Carlson et al., 2008; Fried et al., 2004; Tan et al., 2006; Varma et al., 2016), older

adults who had expressed an interest in volunteering with children were recruited to participate

in a school-based intergenerational program (known as Experience Corps) that involved

supporting children’s literacy development and teaching problem solving and conflict resolution

skills. At 4-8 months follow-up, those who participated in the program were more likely than

those in the waitlist control group to have favorable physical health outcomes (e.g., feeling
 5

stronger and smaller declines in walking speed). A significant group difference was also

observed in the number of people participants felt they could turn to for help. Changes in

psychological health were not assessed.

While these two trials were important in establishing volunteering as a causal predictor of

favorable health outcomes, they were limited in several ways. First, most of the older adults

recruited into these studies had a pre-existing interest in engaging in child-related volunteering

activities and were therefore unlikely to be representative of the broader population of older

adults. Second, as the volunteer activities assessed were limited to interacting with children, the

potential outcomes of participation in a wider range of volunteering activities were not

examined. Third, the amount of volunteering undertaken in each of these trials was substantial,

with participants in the child inpatient hospital visitation program volunteering for 20 hours per

week and those in the intergenerational program volunteering at least 15 hours per week. The

time-intensive nature of the volunteering is likely to have attracted older adults who were

especially motivated to participate at this level and capable of doing so.

Two recent RCTs have addressed many of these limitations. The first, conducted by the

present author team, involved older adults with no known interest in formal volunteering

(blinded for review). Recruiting a broader profile of participants addressed some of the

limitations associated with the prior studies’ sole focus on older people with a pre-stated

intention to undertake volunteering. In addition, those in the intervention condition were able to

choose their own volunteering activities, resulting in a wider range of volunteering roles being

undertaken compared to those of the previous trials. Finally, to better reflect the reality of older

people’s volunteering and the competing demands on their time, those in the volunteering
 6

condition were only required to volunteer for a minimum of 1 hour per week over a 6-month

period.

While some differences in physical health outcomes emerged in our RCT between those

assigned to the intervention and control conditions at 6-month follow-up (blinded for review), the

psychosocial benefits found in previous research were not observed (i.e., higher levels of life

satisfaction; higher self-esteem, personal growth, and purpose in life; lower rates of depression;

greater social connectedness and social support: Carr et al., 2018; Cho et al., 2018; Greenfield &

Marks, 2004; Han & Hong, 2013; Heo et al., 2017; Lum & Lightfoot, 2005; Pilkington et al.,

2012; Wahrendorf et al., 2008). These results were partially supported by the second RCT, in

which older adults prompted to increase their engagement in voluntary work did not report

increased self-efficacy, perceived autonomy, or purpose in life, but did report fewer depressive

symptoms at 6-month follow-up (Jiang et al., 2020). Given it has been suggested that the benefits

of volunteering are most likely to emerge “in the medium to long term, when social networks

and attitudes towards life have had the chance to change” (Russell et al., 2019, p. 119), the 6-

month follow-up period adopted by both these RCTs may have been too short to enable

significant change to occur across many of the assessed psychosocial variables.

The present study aimed to extend our previous research by examining 12-month follow-

up data to determine whether favorable psychosocial outcomes became evident over a longer

time period. The outcomes assessed included depressive symptoms, psychological well-being,

self-esteem, self-efficacy, purpose in life, personal growth, life satisfaction, and social

connectedness.

Method
 7

Design and procedure

This RCT was registered with the Australian and New Zealand Clinical Trial Registry (blinded

for review) and approval to conduct the research was received from a university Human

Research Ethics Committee. The data reported on in the present study were collected via self-

administered surveys at baseline (Time 1: T1) and at 6-month (Time 2: T2) and 12-month (Time

3: T3) follow-up time points.

A parallel-group design was adopted in which all eligible participants (n = 559) were allocated to

either the intervention (n = 280) or control (n = 279) condition based on a computer-generated

script using a simple randomization procedure. Those in the intervention condition were asked to

undertake a minimum of 60 minutes of formal volunteering per week between T1 and T2. The

60-minute threshold was chosen because prior research has identified a curvilinear relationship

between the number of hours spent volunteering and favorable psychological outcomes, with

participation in approximately 100 hours of volunteering per year (up to 2 hours per week)

considered optimal (Morrow-Howell et al., 2003). Those assigned to the control condition were

not asked to volunteer between T1 and T2, but given the potential benefits of participation in this

activity, for ethical reasons they were not advised that they needed to refrain from volunteering.

Participants in both the intervention and control conditions were informed of the requirements of

the study by a member of the study team over the phone and were subsequently given an

information sheet. Participants in both conditions were blinded to the purpose of the study (i.e.,

participants in the control condition were not told about the parallel intervention condition
 8

involving volunteering, and those in the intervention condition were not told about the control

condition that did not involve volunteering).

Recruitment and participants

Various methods were used to recruit community-dwelling older Australians into the trial. As

described elsewhere (blinded for review), these included radio advertising and the placement of

notices in newspapers, retirement villages, and the offices of relevant government authorities. All

advertising materials indicated the study was examining the health and well-being of older

Australians. As per the study protocol (see blinded for review), eligibility criteria were being

aged 60+ years, no engagement in formal volunteering during the previous 12 months, and being

retired. Those in paid employment were ineligible because their participation in the workforce

was likely to provide psychosocial benefits (Maimaris et al., 2010), potentially confounding the

study outcomes.

The demographic profiles of the sample at T1 and T3 are presented in the online

supplementary material (Table S1). At T1, 445 Australian older adults participated in the study.

All provided informed written consent. Participants ranged in age from 60 to 95 years (M =

70.39 years, SD = 6.07), and 56% of the sample was female. By T3, 244 participants remained in

the study (range 61-96 years, M = 71.44 years, SD = 6.09, 55% female, 41% intervention

condition), representing an attrition rate of 45%. This rate of attrition is consistent with other

longitudinal research involving interventions with older adults (e.g., Busetto et al., 2009; Jancey

et al., 2007; Spek et al., 2008). A logistic regression (results of which are presented in the online

supplementary material: Table S2) revealed that attrition rates were equal for gender, age,

socioeconomic status (SES), and condition allocation, but not health status; those in poorer
 9

health were more likely than those in better health to withdraw from the study. A CONSORT

flow diagram of participants’ progress through the phases of the trial is presented in Figure 1.

See blinded for review for further information regarding adherence to CONSORT guidelines.

[Insert Figure 1 about here]

Measures

Psychosocial outcomes. The following measures were used at all three time points to assess

various aspects of psychosocial health. Depressive symptoms were assessed with the 20-item

Center for Epidemiological Studies Depression Scale (Radloff, 1977). Items such as “I felt

depressed” were rated on a scale of 1 (Rarely or none of the time) to 4 (Most or all of the time).

Cronbach’s alphas for scores on this scale at T1, T2, and T3 were 0.87, 0.87, and 0.88

respectively. Psychological well-being was assessed with the 14-item Warwick-Edinburgh

Mental Well-Being Scale (Tennant et al., 2007). Responses to items (e.g., I’ve been feeling

relaxed) were made on a scale of 1 (None of the time) to 5 (All of the time). Cronbach’s alphas of

0.92, 0.93, and 0.93 were obtained at T1, T2, and T3 respectively. The 10-item Rosenberg Self-

Esteem Scale (Rosenberg, 1965) was used to measure self-esteem, with items such as “I feel that

I’m a person of worth, at least on an equal plane with others” rated on a scale of 0 (Strongly

disagree) to 3 (Strongly agree). Cronbach’s alphas for scores on this scale at T1, T2, and T3

were 0.81, 0.89, and 0.88 respectively.

Self-efficacy was assessed using the General Self-Efficacy Scale (Schwarzer &

Jerusalem, 1995). Responses to items such as “No matter what comes my way, I’m usually able

to handle it” were made on a scale of 1 (Not at all true) to 4 (Exactly true), with Cronbach’s
 10

alphas of 0.89, 0.91, and 0.91 obtained at T1, T2, and T3 respectively. Purpose in Life and

Personal Growth were assessed using the relevant subscales of Ryff’s Psychological Well-Being

Scales (Ryff, 1989). Responses to items such as “I have a sense of direction and purpose in life”

and “I am the kind of person who likes to give new things a try” were made on a scale of 1

(Strongly disagree) to 5 (Strongly agree). Cronbach’s alphas of 0.90, 0.88, and 0.90 were

obtained at T1, T2, and T3 respectively for Purpose in Life. Cronbach’s alphas of 0.86, 0.86, and

0.88 were obtained at T1, T2, and T3 respectively for Personal Growth. Life satisfaction was

assessed using a single-item scale adapted from Van Willigen (2000): “All things considered,

how satisfied are you with your life as a whole these days?” (Response options: 1 = Very good to

5 = Very bad, reverse-scored). Finally, social connectedness was measured using the 24-item

Social Provisions Scale (Cutrona & Russell, 1987). Items (e.g., There is someone I could talk to

about important decisions in my life) were rated on a scale that ranged from 1 (Strongly

disagree) to 4 (Strongly agree). Cronbach’s alphas for scores on this scale at T1, T2, and T3

were 0.91, 0.92, and 0.94 respectively.

Volunteering. At T2 and T3, participants reported whether they had engaged in formal

volunteering in the previous 6 months (yes/no response options). Those responding in the

affirmative were asked to report the number of organizations for which they had volunteered and

the average number of hours per week they engaged in volunteering. Responses to these

questions at T2 were used to determine compliance with the conditions of the RCT. Those in the

intervention condition who responded ‘yes’ to engaging in volunteering in the previous 6

months, had volunteered for at least one organization, and had volunteered for a minimum of 60

minutes per week were deemed compliant, as were those in the control condition who responded
 11

‘no’ to engaging in volunteering. By contrast, those in the intervention condition who responded

‘no’ to engaging in volunteering in the previous 6 months, had not volunteered for an

organization, or had volunteered for less than 60 minutes per week were deemed non-compliant,

as were those in the control condition who responded ‘yes’ to volunteering. Responses at T3

were used to generate descriptive data on the proportion of participants in the intervention

condition who chose to continue volunteering beyond the prescribed intervention period.

Socio-demographics. Participants were asked to report on various socio-demographic

characteristics (e.g., gender, age, highest level of education, postcode) at all three time points.

Postcodes were used to calculate the SES of the area in which participants resided (as per the

Australian Bureau of Statistics’ Socioeconomic Index for Areas: Australian Bureau of Statistics,

2018).

Analysis

As results from the 6-month follow-up have been reported elsewhere (blinded for review),

analyses were conducted comparing baseline to 12-month follow-up data only. A per-protocol

approach was adopted such that participants who provided T1 data but did not complete the T2

or T3 assessments were excluded from analyses. Comparisons were made between those who

complied with the intervention condition (i.e., met the minimum 60 minutes per week

requirement for volunteering between T1 and T2) and those who complied with the control

condition (i.e., no volunteering between T1 and T2). Sensitivity analyses were also conducted in

the form of pragmatic analyses. These compared outcomes for those participants who engaged in
 12

formal volunteering to those of participants who did not engage in volunteering, regardless of

assigned condition.

Paired samples t-tests were used to assess changes on the psychosocial outcome variables

listed above between T1 and T3 within groups. Hierarchical linear regression analyses were

conducted to compare differences between groups. The T1 score of the psychosocial variable

under investigation was entered in Step 1 of the analysis. As a gender imbalance was observed

by condition assignment owing to the simple rather than stratified randomization procedure

adopted, gender was also entered in Step 1 as a control variable (imbalances were not observed

for age, SES, or objective health status). The group variable was entered in Step 2. As only 244

participants remained in the sample at T3, bootstrapping (n = 1000 replications) was used in

estimations to reduce the effect of any excess variability (as per Nevitt & Hancock, 2001). All

analyses were conducted in SPSS.

Among those participants compliant with the volunteering (n = 106) and control (n =

178) conditions at T2, n = 73 and n = 112 participants respectively remained in the sample at T3.

The demographic profile of T3 participants stratified by condition is provided in the online

supplementary material (Table S3). Owing to non-response on the items assessing volunteering

by some participants, engagement in volunteering between T2 and T3 was only able to be

assessed for 160 of these 185 participants. Of these, 41 (66%) of those who had been compliant

with the volunteering condition at T2 continued their engagement in volunteering between T2

and T3, while 88 (90%) of participants compliant with the control condition at T2 remained non-

volunteers. Given these small sample sizes, analyses of the impact of continued volunteering

between T2 and T3 were not feasible.

 13

Results

Outcomes by allocated condition: per-protocol analyses

Scores on each of the psychosocial variables under investigation at T1 and T3 stratified by group

are presented in Table 1. Despite scores on the outcome variables being moderate to high at T1,

significant improvements over time were observed among those in the intervention arm who

complied with the study requirements and engaged in volunteering on the variables of life

satisfaction (p < .001), social connectedness (p < .001), purpose in life (p = .001), self-esteem (p

= .014), and personal growth (p = .028). Improvements over time for two of the outcomes were

also observed among those in the control arm who did not engage in volunteering: self-esteem (p

= .001) and social connectedness (p = .008).

[Insert Table 1 about here]

Step 2 results from the per-protocol hierarchical regression analyses are presented in

Table 2 (Step 1 results are shown in Table S4 in the online supplementary material). Three

significant between-group differences emerged: those in the intervention arm who complied with

the study requirements and engaged in volunteering between T1 and T2 demonstrated significant

improvements in purpose in life (p = .016), personal growth (p = .020), and life satisfaction (p =

.020) scores by T3 (while controlling for T1 scores and gender) relative to those in the control

arm who did not engage in volunteering.

[Insert Table 2 about here]

 14

Outcomes by actual volunteering status: pragmatic analyses

Table 3 presents T1 and T3 scores stratified by actual volunteering status (i.e., irrespective of

allocated condition). Significant improvements were observed on all variables among those who

volunteered between T1 and T2. Among those who did not volunteer, a significant improvement

was observed for the variables of self-esteem and social connectedness only (see Table 3). One

significant between-group difference emerged: those who engaged in volunteering between T1

and T2 demonstrated significant improvements in personal growth (p = .031) scores at T3 (while

controlling for T1) relative to those who did not engage in volunteering.

[Insert Table 3 about here]

Discussion

This study examined 12-month follow-up data as part of an RCT examining the effects of

commencing volunteering on the well-being of older adult non-volunteers (blinded for review).

Per-protocol analyses indicated that those in the intervention arm who complied with the study

requirements and engaged in volunteering reported significant improvements on the variables of

life satisfaction, purpose in life, and personal growth compared to those in the control arm who

did not engage in volunteering. Pragmatic analyses indicated that those who volunteered reported

significant improvements in personal growth compared to those who did not volunteer,

regardless of the condition to which they had been allocated. Outcomes on these psychological

variables were not found to differ significantly between groups at the 6-month follow-up

(blinded for review), suggesting the impact of volunteering on these outcomes may take some

time to manifest and that benefits are more likely to be observed in the medium to long term.
 15

Given results at 6-month follow-up were trending in the expected direction, it is possible the

benefits of volunteering began to emerge in the short term but were not at the level required to

reach statistical significance.

These results support the findings of previous cross-sectional and longitudinal research

comparing older adult volunteers and non-volunteers (e.g., Cho et al., 2018; Greenfield & Marks,

2004; Heo et al., 2017; Van Willigen, 2000) by providing evidence of a causal relationship

between engagement in volunteering and the variables of life satisfaction, purpose in life, and

personal growth. The significant improvements in these outcomes observed among those in the

intervention condition compared to those in the control condition also provide support for

research that has suggested engagement in volunteering may protect older adults from the

reductions in purpose in life and personal growth that can accompany aging due to the loss of

major roles such as workforce participation (Greenfield & Marks, 2004; Heo et al., 2017),

thereby facilitating healthy aging (Lemon et al., 1972).

Volunteering was found to have no significant effect on psychological well-being,

depression, self-efficacy, self-esteem, or social connectedness. This outcome differs from prior

longitudinal and cross-sectional research that found an association between volunteering and

these factors (Han & Hong, 2013; Haski-Leventhal, 2009; Li & Ferraro, 2005; Morrow-Howell

et al., 2003; Musick & Wilson, 2003). In terms of the results of prior RCTs, the non-significant

findings for self-esteem and depression observed in the present study are consistent with

outcomes from Rook and Sorkin (2003), but the finding related to depression is inconsistent with

the results of Jiang et al. (2020). Further research exploring these mixed results is warranted.

Implications
 16

The results of the present study have several implications. First, they indicate that volunteering

can have important health benefits, providing support for the suggestion that engagement in

social and/or productive activities in general has the potential to promote healthy aging (Lemon

et al., 1972; Rowe & Kahn, 1997), and that the ‘prescription’ of volunteering in particular could

constitute an effective healthy aging intervention (Pettigrew et al., 2019). However, it appears

that psychological benefits may take some time to manifest, highlighting the need to ensure that

older recruits are informed of the likely time periods for these personal benefits to accrue. For

example, medical practitioners could recommend volunteering to their patients and explain the

time frames over which different kinds of outcomes could be expected. The results of the present

RCT suggest that improvements in physical outcomes may occur first (blinded for review),

followed later by psychological benefits. Such an approach has the potential to enhance

volunteer retention, thereby improving outcomes for individuals and society. It is encouraging

that of those participants in the intervention condition who complied with the study requirements

and engaged in volunteering between T1 and T2, two-thirds chose to continue volunteering after

the 6-month trial period was complete and were still volunteering at T3.

Second, the outcomes of this study suggest changes that could be made to the design of future

RCTs examining the psychosocial benefits of volunteering. Specifically, it seems prudent to

ensure these RCTs are designed to incorporate a follow-up period of at least 12 months to enable

detection of changes that take some time to manifest at significant levels. Finally, the loss of a

substantial minority of participants to follow-up highlights the importance of designing RCTs in

such a way that minimizes attrition. Participants’ reasons for withdrawing from the present study

were not collected systematically; however, anecdotal reports from those who did provide
 17

feedback indicated competing commitments and the emergence or exacerbation of health

problems were primary reasons. In terms of the former, many of those who provided reasons for

their withdrawal reported that they did not have time to participate given their other

responsibilities and lifestyle choices, which included (i) providing care to their grandchildren and

ill spouses and relatives; (ii) attending meetings or programs of groups, clubs, and organizations

to which they belonged (e.g., Probus, University of the Third Age, church groups); and (iii)

engaging in spontaneous travel. These tentative findings suggest that researchers designing

volunteering RCTs and health practitioners recommending volunteering to patients should be

mindful of relevant barriers and assist older people to select activities that can accommodate

health conditions, other responsibilities, and busy schedules.

Limitations

The present study had several limitations. First, as noted above, a substantial minority of

participants was lost to follow-up. Although the observed attrition rate is consistent with prior

longitudinal research involving interventions with older adults (e.g., Busetto et al., 2009; Jancey

et al., 2007; Spek et al., 2008), the sample sizes by condition were too small to permit an

assessment of the extent to which continued volunteering between T2 and T3 impacted on the

observed outcomes. Second, although attrition rates were equal for gender, age, SES, and

condition allocation, participants lost to follow-up had poorer health and may have differed from

those who remained at T3 on factors that were not measured in the present study. Third, baseline

scores on the outcome variables were high. This suggests the sample was relatively

psychologically healthy, reducing the generalizability of the results to older people in general. In

addition, high baseline scores are likely to have made it difficult to observe significant
 18

improvements over time. Despite this, some statistically significant improvements in

psychological health were observed, attesting to the robustness of the findings.

Fourth, the potential impact of the type of volunteering in which participants engaged could not

be explored as there was substantial variation in the volunteering roles adopted across the

sample, resulting in a lack of statistical power. Research conducted in a larger sample is needed

to examine the effect of volunteering type on psychosocial well-being. Finally, the recruitment

procedures adopted for the study required participants to self-select, potentially resulting in bias.

In addition, all the participants in this follow-up study were volunteers to the extent that they

opted to participate, and continue participating, in the study. Research is needed to explore the

outcomes associated with commencing volunteering among the broader population of older

people, especially among those who are less likely to volunteer for an extended period of time.

Conclusion

Volunteering commencement was found to be associated with significant increases in older

adults’ life satisfaction, purpose in life, and personal growth over a 12-month period. These

findings provide causal support for the benefits of volunteering among older adults and indicate

that encouraging participation in this activity could constitute an effective healthy aging

intervention.
 19

References

Anderson, N. D., Damianakis, T., Kröger, E., Wagner, L. M., Dawson, D. R., Binns, M. A.,
Bernstein, S., Caspi, E., & Cook, S. L. (2014). The benefits associated with volunteering
among seniors: A critical review and recommendations for future research. Psychological
Bulletin, 140(6), 1505-1533. https://doi.org/10.1037/a0037610
Australian Bureau of Statistics. (2018). Census of Population and Housing: Socio-Economic
Indexes for Areas (SEIFA), Australia, 2016 (Cat. no. 2033.0.55.001). Canberra: ABS
Burr, J. A., Han, S. H., & Tavares, J. L. (2015). Volunteering and cardiovascular disease risk:
Does helping others get “under the skin?”. The Gerontologist, 56(5), 937-947.
https://doi.org/10.1093/geront/gnv032
Busetto, L., Mazza, M., Salvalaio, S., De Stefano, F., Marangon, M., Calò, E., Sampietro, S., &
Enzi, G. (2009). Obesity treatment in elderly outpatients: predictors of efficacy and drop-
out. Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity, 14, e56-
e65. https://doi.org/https://doi.org/10.1007/BF03327801
Carlson, M. C., Saczynski, J. S., Rebok, G. W., Seeman, T., Glass, T. A., McGill, S., Tielsch, J.,
Frick, K. D., Hill, J., & Fried, L. P. (2008). Exploring the effects of an “everyday”
activity program on executive function and memory in older adults: Experience Corps®.
The Gerontologist, 48, 793-801. https://doi.org/10.1093/geront/48.6.793
Carr, D. C., Kail, B. L., Matz-Costa, C., & Shavit, Y. Z. (2018). Does becoming a volunteer
attenuate loneliness among recently widowed older adults? The Journals of Gerontology
Series B: Psychological Sciences and Social Sciences, 73(3), 501-510.
https://doi.org/10.1093/geronb/gbx092
Cho, D., Post, J., & Kim, S. K. (2018). Comparison of passive and active leisure activities and
life satisfaction with aging. Geriatrics & Gerontology International, 18(3), 380-386.
https://doi.org/10.1111/ggi.13188
Cutrona, C. E., & Russell, D. W. (1987). The provisions of social relationships and adaptation to
stress. Advances in Personal Relationships, 1(1), 37-67.
Fried, L. P., Carlson, M. C., Freedman, M., Frick, K. D., Glass, T. A., Hill, J., McGill, S., Rebok,
G. W., Seeman, T., Tielsch, J., Wasik, B. A., & Zeger, S. (2004). A social model for
health promotion for an aging population: Initial evidence on the Experience Corps
model. Journal of Urban Health, 81(1), 64-78. https://doi.org/10.1093/jurban/jth094
Greenfield, E. A., & Marks, N. F. (2004). Formal volunteering as a protective factor for older
adults' psychological well-being. The Journals of Gerontology Series B: Psychological
Sciences and Social Sciences, 59(5), S258-S264.
https://doi.org/10.1093/geronb/59.5.S258
Han, C. K., & Hong, S. I. (2013). Trajectories of volunteering and self-esteem in later life: Does
wealth matter? Research on Aging, 35(5), 571-590.
https://doi.org/10.1177/0164027512449472
Haski-Leventhal, D. (2009). Elderly volunteering and well-being: A cross-European comparison
based on SHARE data. VOLUNTAS: International Journal of Voluntary and Nonprofit
Organizations, 20(4), 388-404. https://doi.org/10.1007/s11266-009-9096-x
Heo, J., Chun, S., Kim, B., Ryu, J., & Lee, Y. (2017). Leisure activities, optimism, and personal
growth among the young-old, old-old, and oldest-old. Educational Gerontology, 43(6),
289-299. https://doi.org/10.1080/03601277.2017.1289457
 20

Hong, S. I., & Morrow-Howell, N. (2010). Health outcomes of Experience Corps: A high-
commitment volunteer program. Social Science & Medicine. Part A: Medical Sociology,
71, 414-420. https://doi.org/10.1016/j.socscimed.2010.04.009

Jancey, J., Lee, A., Howat, P., Clarke, A., Wang, K., & Shilton, T. (2007). Reducing attrition in
physical activity programs for older adults. Journal of Aging and Physical Activity, 15,
152-165.
Jenkinson, C. E., Dickens, A. P., Jones, K., Thompson-Coon, J., Taylor, R. S., Rogers, M.,
Bambra, C. L., Lang, I., & Richards, S. H. (2013). Is volunteering a public health
intervention? A systematic review and meta-analysis of the health and survival of
volunteers. BMC Public Health, 13. https://doi.org/10.1186/1471-2458-13-773
Jiang, D., Warner, L. M., Chong, A. M. L., Li, T., Wolff, J. K., & Chou, K. L. (2020). Benefits
of volunteering on psychological well-being in older adulthood: Evidence from a
randomized controlled trial. Aging & Mental Health.
https://doi.org/10.1080/13607863.2020.1711862
Lemon, B. W., Bengtson, V. L., & Peterson, J. A. (1972). An exploration of the activity theory
of aging: Activity types and life satisfaction among in-movers to a retirement community.
Journal of Gerontology, 27(4), 511-523. https://doi.org/10.1093/geronj/27.4.511
Li, Y., & Ferraro, K. F. (2005). Volunteering and depression in later life: Social benefit or
selection processes? Journal of Health and Social Behavior, 46(1), 68-84.
https://doi.org/10.1177/002214650504600106
Lum, T. Y., & Lightfoot, E. (2005). The effects of volunteering on the physical and mental
health of older people. Research on Aging, 27(1), 31-55.
https://doi.org/10.1177/0164027504271349
Maimaris, W., Hogan, H., & Lock, K. (2010). The impact of working beyond traditional
retirement ages on mental health: implications for public health and welfare policy.
Public Health Reviews, 32(2), 532-548.
https://doi.org/https://doi.org/10.1007/BF03391615
Morrow-Howell, N., Hinterlong, J., Rozario, P. A., & Tang, F. (2003). Effects of volunteering on
the well-being of older adults. The Journals of Gerontology Series B: Psychological
Sciences and Social Sciences, 58(3), S137-S145.
https://doi.org/10.1093/geronb/58.3.S137
Musick, M. A., & Wilson, J. (2003). Volunteering and depression: The role of psychological and
social resources in different age groups. Social Science and Medicine, 56(2), 259-269.
https://doi.org/10.1016/S0277-9536(02)00025-4
Nevitt, J., & Hancock, G. R. (2001). Performance of bootstrapping approaches to model test
statistics and parameter standard error estimation in structural equation modeling.
Structural Equation Modeling, 8(3), 353-377.
https://doi.org/10.1207/S15328007SEM0803_2
Parkinson, L., Warburton, J., Sibbritt, D., & Byles, J. (2010). Volunteering and older women:
Psychosocial and health predictors of participation. Aging & Mental Health, 14(8), 917-
927. https://doi.org/10.1080/13607861003801045
Pettigrew, S., Jongenelis, M. I., Jackson, B., Warburton, J., & Newton, R. U. (2019). A
randomized controlled trial and pragmatic analysis of the effects of volunteering on the
health and well-being of older people. Aging Clinical and Experimental Research.
https://doi.org/10.1007/s40520-019-01241-3
 21

Pettigrew, S., & Roberts, M. (2008). Addressing loneliness in later life. Aging and Mental
Health, 12(3), 302-309. https://doi.org/https://doi.org/10.1080/13607860802121084
Pilkington, P. D., Windsor, T. D., & Crisp, D. A. (2012). Volunteering and subjective well-being
in midlife and older adults: The role of supportive social networks. The Journals of
Gerontology Series B: Psychological Sciences and Social Sciences, 67(2), 249-260.
https://doi.org/10.1093/geronb/gbr154
Proulx, C. M., Curl, A. L., & Ermer, A. E. (2018). Longitudinal associations between formal
volunteering and cognitive functioning. The Journals of Gerontology Series B:
Psychological Sciences and Social Sciences, 73(3), 522-531.
https://doi.org/10.1093/geronb/gbx110
Radloff, L. (1977). The CES-D Scale: A self-report depression scale for research in the general
population. Applied Psychological Measurement, 1(3), 385-401.
https://doi.org/10.1177/014662167700100306
Rook, K. S., & Sorkin, D. H. (2003). Fostering social ties through a volunteer role: Implications
for older-adults' psychological health. The International Journal of Aging and Human
Development, 57(4), 313-337. https://doi.org/10.2190/NBBN-EU3H-4Q1N-UXHR
Rosenberg, M. (1965). Society and the adolescent self-image. Princeton University Press.
Rowe, J. W., & Kahn, R. L. (1997). Successful aging. The Gerontologist, 37(4), 433-440.
https://doi.org/10.1093/geront/37.4.433
Russell, A. R., Nyame-Mensah, A., de Wit, A., & Handy, F. (2019). Volunteering and wellbeing
among ageing adults: A longitudinal analysis. VOLUNTAS: International Journal of
Voluntary and Nonprofit Organizations, 30(1), 115-128. https://doi.org/10.1007/s11266-
018-0041-8
Ryff, C. D. (1989). Happiness is everything, or is it? Explorations on the meaning of
psychological well-being. Journal of Personality and Social Psychology, 57(6), 1069-
1081. https://doi.org/10.1037/0022-3514.57.6.1069
Schwarzer, R., & Jerusalem, M. (1995). Generalized Self-Efficacy scale. In J. Weinman, S.
Wright, & M. Johnston (Eds.), Measures in health psychology: A user’s portfolio (pp. 35-
37). NFER-Nelson.
Spek, V., Cuijpers, P., Nyklicek, I., Smits, N., Riper, H., Keyzer, J., & Pop, V. (2008). One-year
follow-up results of a randomized controlled clinical trial on internet-based cognitive
behavioural therapy for subthreshold depression in people over 50 years. Psychological
Medicine, 38(5), 1797–1806. https://doi.org/10.1017/S0033291707000542
Tan, E. J., Rebok, G. W., Yu, Q., Frangakis, C. E., Carlson, M. C., Wang, T., Ricks, M., Tanner,
E. K., McGill, S., & Fried, L. P. (2009). The long-term relationship between high-
intensity volunteering and physical activity in older African American women. The
Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 64(2),
304-311.
Tan, E. J., Xue, Q. L., Li, T., Carlson, M. C., & Fried, L. P. (2006). Volunteering: A physical
activity intervention for older adults—the experience Corps® program in Baltimore.
Journal of Urban Health, 83(5), 954-969. https://doi.org/https://doi.org/10.1007/s11524-
006-9060-7
Tennant, R., Hiller, L., Fishwick, R., Platt, S., Joseph, S., Weich, S., Parkinson, J., Secker, J., &
Stewart-Brown, S. (2007). The Warwick-Edinburgh Mental Well-Being Scale
(WEMWBS): Development and UK validation. Health and Quality of Life Outcomes, 5.
https://doi.org/10.1186/1477-7525-5-63
 22

Tomioka, K., Kurumatani, N., & Hosoi, H. (2017). Association between the frequency and
autonomy of social participation and self‐rated health. Geriatrics & Gerontology
International, 17(12), 2537-2544. https://doi.org/10.1111/ggi.13074
Van Willigen, M. (2000). Differential benefits of volunteering across the life course. The
Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 55(5),
S308-S318. https://doi.org/10.1093/geronb/55.5.S308
Varma, V. R., Tan, E. J., Gross, A. L., Harris, G., Romani, W., Fried, L. P., Rebok, G. W., &
Carlson, M. C. (2016). Effect of community volunteering on physical activity: A
randomized controlled trial. American Journal of Preventive Medicine, 50(1), 106-110.
https://doi.org/10.1016/j.amepre.2015.06.015
Wahrendorf, M., Ribet, C., Zins, M., & Siegrist, J. (2008). Social productivity and depressive
symptoms in early old age–results from the GAZEL study. Aging and Mental Health,
12(3), 310-316. https://doi.org/https://doi.org/10.1080/13607860802120805
World Health Organization. (2015). World report on ageing and health. Geneva: WHO
 23

Table 1

Outcomes by Allocated Condition (n = 185): Per-Protocol Analyses

Measures Volunteer condition: complianta (n = 73) Control condition: compliantb (n = 112)

T1 T3 T1-T3 T1 T3 T1-T3

M SD M SD  d M SD M SD  d

Psychological well-being (range: 14 – 70) 56.07 7.97 57.76 7.61 1.69 0.21 54.36 7.80 55.01 8.42 0.64 0.09

Depression (range: 0 – 60) 8.39 7.65 6.90 6.75 -1.49 -0.22 7.49 7.71 7.69 7.80 0.21 0.03

Self-efficacy (range: 10 – 40) 33.29 3.97 33.99 4.49 0.70 0.22 31.88 3.96 32.47 4.39 0.59 0.18

Purpose in life (range: 14 – 70) 67.13 12.66 70.84 10.81 3.70** 0.42 66.58 12.01 67.14 11.78 0.55 0.07

Personal growth (range: 14 – 70) 70.38 10.39 73.28 9.21 2.90* 0.27 68.41 9.10 68.48 9.89 0.06 0.01

Self-esteem (range: 0 – 30) 23.55 4.90 24.97 4.34 1.42* 0.31 23.68 5.47 25.08 4.57 1.41** 0.33

Life satisfaction (range: 1 – 5) 4.07 0.77 4.36 0.64 0.29*** 0.42 4.08 0.79 4.14 0.75 0.06 0.07

Social connectedness (range: 24 – 96) 79.79 10.26 82.89 10.92 3.11*** 0.47 79.02 9.14 81.20 10.14 2.18** 0.27

Note. Those who completed their T3 assessment but did not provide data relating to one or more of the outcome variables were treated listwise.
a
Responded ‘yes’ to engaging in volunteering in the previous 6 months, had volunteered for at least one organization, and had volunteered for an average of 60

minutes per week.

b
Responded ‘no’ to engaging in volunteering.

*p < .05. **p < .01. ***p < .001.

 24

Table 2

Per-Protocol Hierarchical Regression Analyses (n = 185)

Dependent variable Independent variable B SE β p 95% CI

T3 psychological well-being Gender 0.78 1.12 .05 .504 -1.44, 2.93
T1 psychological well-being 0.59 0.10 .57 <. 001 0.40, 0.79
Group 1.58 1.11 .09 .159 -0.56, 3.81
T3 depression Gender -0.88 1.03 -.06 .403 -2.87, 1.21
T1 depression 0.53 0.10 .54 < .001 0.33, 0.72
Group -1.06 1.04 -.07 .298 -3.23, 1.02
T3 self-efficacy Gender 1.01 0.46 .11 .035 0.13, 2.01
T1 self-efficacy 0.78 0.06 .70 < .001 0.65, 0.90
Group 0.22 0.49 -.02 .647 -0.81, 1.13
T3 purpose in life Gender 0.50 1.29 .02 .695 -1.93, 3.04
T1 purpose in life 0.69 0.08 .73 < .001 0.54, 0.84
Group 3.22 1.25 .14 .016 0.62, 5.50
T3 personal growth Gender 2.04 1.22 .10 .106 -0.28, 4.46
T1 personal growth 0.58 0.09 .56 < .001 0.41, 0.76
Group 3.26 1.28 .16 .020 0.52, 5.80
T3 self-esteem Gender 0.70 0.54 .08 .192 -0.41, 1.79
T1 self-esteem 0.51 0.09 .60 < .001 0.35, 0.69
Group -0.21 0.59 -.02 .732 -1.37, 0.96
T3 life satisfaction Gender -0.02 0.11 -.01 .869 -0.24, 0.21
T1 life satisfaction 0.42 0.09 .46 < .001 0.26, 0.60
Group 0.23 0.10 .16 .020 0.04, 0.42
T3 social connectedness Gender 1.43 1.21 .07 .251 -0.90, 3.82
T1 social connectedness 0.77 0.06 .70 < .001 0.65, 0.88
Group 0.82 1.10 .04 .468 -1.44, 2.96
Note. Significant group effect shown in bold text. β = not bootstrapped. Group: 1 = compliant with control condition, 2 = compliant with intervention condition.
Gender: 1 = male, 2 = female. Step 1 of hierarchical regression analysis not presented.
 25

Table 3

Outcomes by Volunteering Status, Irrespective of Allocated Condition (n = 229a): Pragmatic Analyses

Measures Volunteered between T1 and T2 (n = 101) Did not volunteer between T1 and T2 (n = 128) Regression results
T1 T3 T1-T3 T1 T3 T1-T3
M SD M SD  d M SD M SD  d
Psychological well-being 55.74 7.85 57.81 8.14 2.06* 0.26 54.67 7.64 55.34 8.31 0.67 0.10 B = -1.77, SE = 0.95, β = -.11, p = .069,
95% CI = -3.59, 0.17
Depression 8.11 7.19 6.42 6.50 -1.69* -0.28 7.37 7.35 7.37 7.44 0.00 0.00 B = 1.28, SE = 0.83, β = .09, p = .125,
95% CI = -0.34, 2.88
Self-efficacy 33.19 3.79 34.00 4.34 0.81* 0.25 32.09 4.01 32.68 4.41 0.59 0.18 B = -0.38, SE = 0.42, β = -.04, p = .580,
95% CI = -5.11, 9.04
Purpose in life 67.46 12.38 69.93 10.70 2.46* 0.27 67.08 11.65 67.87 11.52 0.79 0.10 B = -1.66, SE = 1.11, β = -.07, p = .152,
95% CI = -3.87, 0.56
Personal growth 70.94 10.20 72.98 9.38 2.04* 0.20 68.56 9.24 68.97 9.62 0.41 0.05 B = -2.44, SE = 1.08, β = -.13, p = .031,
95% CI = -4.43, -0.23
Self-esteem 23.61 4.98 25.05 4.63 1.44** 0.31 23.92 5.31 25.25 4.43 1.33*** 0.33 B = 0.12, SE = 0.53, β = .01, p = .822,
95% CI = -0.94, 1.15
Life satisfaction 4.03 0.81 4.26 0.74 0.22** 0.29 4.11 0.76 4.17 0.73 0.06 0.07 B = -0.12, SE = 0.09, β = -.08, p = .201,
95% CI = -0.29, 0.07
Social connectedness 79.25 10.57 82.40 10.50 3.15*** 0.46 79.31 9.05 81.61 9.90 2.30** 0.29 B = -0.74, SE = 0.99, β = -.04, p = .451,
95% CI = -2.65, 1.19
Note. Those who completed their T3 assessment but did not provide data relating to one or more of the outcome variables were treated listwise. Significant group
effect shown in bold text. Gender treated as a covariate. β = not bootstrapped.
*p < .05. **p < .01. ***p < .001.
a
Volunteering status was unable to be determined for 15 participants (excluded from analyses).
 26

Eligible
N = 559

Withdrew Assigned to intervention Assigned to control Withdrew

n = 79 (28%) n = 280 n = 279 n = 35 (13%)

Assessed at T1 Assessed at T1
n = 201 n = 244
Measures: Measures:
Objective physical health Objective physical health
Subjective physical health Subjective physical health
Psychosocial health Psychosocial health
Withdrew Withdrew
n = 53 n = 22
Assessed at T2 Assessed at T2
n = 148 n = 222
Measures: Measures:
Objective physical health Objective physical health
Subjective physical health Subjective physical health
Psychosocial health Psychosocial health
Lost to follow-up Lost to follow-up
n = 48 n = 78
Assessed at T3 Assessed at T3
n = 100 n = 144
Measures: Measures:
Compliant Compliant
Subjective physical health Subjective physical health
n = 73 Psychosocial health Psychosocial health n = 112
 27

Figure 1
CONSORT Diagram Depicting Progress Through the Study and the Measures Collected at Each Time Point.
Minerva Access is the Institutional Repository of The University of Melbourne

Author/s:
Jongenelis, M;Jackson, B;Newton, RU;Pettigrew, S

Title:
Longitudinal associations between formal volunteering and well-being among retired older
people: follow-up results from a randomized controlled trial

Date:
2021-02-04

Citation:
Jongenelis, M., Jackson, B., Newton, R. U. & Pettigrew, S. (2021). Longitudinal associations
between formal volunteering and well-being among retired older people: follow-up results
from a randomized controlled trial. AGING & MENTAL HEALTH, 26 (2), pp.368-375. https://
doi.org/10.1080/13607863.2021.1884845.

Persistent Link:
http://hdl.handle.net/11343/297104